Production delays hit Business Central Tuesday. Your plant head finds out Friday.
The signal is in BC. Production order at risk, stockout before the next run, cost variance 18% above plan. It has no owner. It has no path to a decision. It surfaces in a report that nobody reads until Thursday.
Book a live OpsGrid demo with manufacturing data →Three signals in Business Central right now: none of them have an owner
OpsGrid monitors BC continuously. When a threshold is crossed, it ranks the signal by cost impact and sends it as an actionable card to the named owner in Teams.
- Material shortage: fabric component below safety stock
- 2 production days remain before customer commit
- Exposure: $22,400 order + £8K OTIF penalty
- 4 days stock remaining at current sales velocity
- Next production run: 12 days out
- Exposure: $5,900 in backorders accumulating daily
- Material substitution approved but not updated in BC
- Variance widening for 9 days, no owner assigned
- Exposure: £3,200 margin erosion this month
Signal to decision in Teams, not in a Friday report
BC crosses a threshold: production at risk, material short, cost variance. OpsGrid detects it and ranks by cost impact.
Plant Head gets production risk. Production Manager gets schedule conflicts. Procurement Lead gets material shortage. One person per signal. SLA clock starts.
Owner reviews the recommended action in Teams (no BC login, no new app). Approve, reject, or escalate with one tap.
Approved action writes to BC: draft purchase order, rescheduled production order, transfer order. Nothing posts without explicit approval.
Every decision, override, and outcome logged. Who decided, when, what BC data they saw, what happened next.
BC Wave 1 2026 shipped native agents. Does OpsGrid still add value?
Microsoft's Wave 1 2026 shipped task-specific agents: Payables Agent handles invoice matching, Sales Order Agent handles order entry. These are single-task tools. They don't watch production signals. They don't route across signal types. They have no governance layer and no audit trail for operational decisions.
OpsGrid operates across the signals that matter to manufacturing: production at risk, stockout before a run, cost variance widening, procurement lag. It assigns each signal to a named owner, captures the decision, and maintains the trail.
The native agents automate task execution. OpsGrid governs how decisions get made across them. They're complementary. You need both or you have automation without accountability.
What mid-market manufacturers ask before deployment
Does OpsGrid replace our ERP or production planning system?
No. OpsGrid connects to Business Central using a service account your BC admin configures with appropriate permissions. It surfaces signals that already exist in BC (production orders at risk, stockout warnings, cost variances) and routes them to named decision owners in Teams with a recommended action. The ERP stays the system of record. OpsGrid is the governance layer that ensures BC data reaches the right person before the cost lands.
How does OpsGrid handle the Microsoft Wave 1 2026 native agents in Business Central?
Microsoft's Wave 1 2026 shipped task-specific agents: Payables Agent handles invoice matching, Sales Order Agent handles order entry. These are single-task tools with no cross-signal view and no governance layer. OpsGrid operates across signal types (production, inventory, procurement, cost) and routes decisions to named owners with audit trail. The two are complementary: native agents automate task execution, OpsGrid governs how decisions get made across them.
What does OpsGrid actually write to Business Central?
Nothing without explicit approval. When an operational signal surfaces (a production order at risk, a purchase order needed), OpsGrid drafts the corrective action and routes it to the named owner in Teams. The owner approves or rejects in one click. Only after approval does OpsGrid write to BC: draft purchase order, transfer order, or rescheduled production order. The architectural constraint means zero unauthorized writes, not a policy that can be toggled.
How long does deployment take for a mid-market manufacturer?
Three weeks. Week 1: BC APIs enabled, service account provisioned, decision types configured to your operation: production signals, procurement thresholds, cost variance limits. Week 2: Teams webhook live, approval routing configured. Week 3: Go-live walkthrough with plant and production team, first live signals reviewed, routing adjustments complete. Measurable ROI within 60 days of go-live or we refund 50% of implementation cost.
See your BC production signals routed to the right person in one session
We start with a free Decision Latency Audit: map where your BC data is getting stuck and put a number on it. No commitment until you see the cost. Dynamics 365 Business Central required. 3-week deployment. ROI in 60 days or 50% refund.